Update README.md
Browse files
README.md
CHANGED
|
@@ -10,15 +10,24 @@ tags:
|
|
| 10 |
- ML
|
| 11 |
- Ai
|
| 12 |
---
|
| 13 |
-
|
|
|
|
| 14 |
Model Information
|
|
|
|
| 15 |
Name: Facial Means
|
|
|
|
| 16 |
Type: Convolutional Neural Network (CNN)
|
|
|
|
| 17 |
Framework: TensorFlow
|
|
|
|
| 18 |
Dataset: Celebrity Faces Dataset
|
|
|
|
| 19 |
Dataset Path: /content/drive/MyDrive/beard_dataset/celb_dataset/
|
|
|
|
| 20 |
Model Save Path: /content/drive/MyDrive/beard_dataset/celebrity_model.h5
|
|
|
|
| 21 |
Image Dimensions: 224 x 224 pixels
|
|
|
|
| 22 |
Batch Size: 32
|
| 23 |
Data Augmentation
|
| 24 |
Rescale: 1./255
|
|
@@ -27,7 +36,8 @@ Zoom Range: 0.2
|
|
| 27 |
Horizontal Flip: True
|
| 28 |
Validation Split: 20%
|
| 29 |
Model Architecture
|
| 30 |
-
Layer (type)
|
|
|
|
| 31 |
===============================================================
|
| 32 |
conv2d (Conv2D) (None, 222, 222, 32) 896
|
| 33 |
max_pooling2d (MaxPooling2D) (None, 111, 111, 32) 0
|
|
@@ -39,6 +49,7 @@ flatten (Flatten) (None, 86528) 0
|
|
| 39 |
dense (Dense) (None, 128) 11075712
|
| 40 |
dense_1 (Dense) (None, 6) 774
|
| 41 |
===============================================================
|
|
|
|
| 42 |
Total params: 11,170,734
|
| 43 |
Trainable params: 11,170,734
|
| 44 |
Non-trainable params: 0
|
|
@@ -46,8 +57,11 @@ Model Compilation
|
|
| 46 |
Optimizer: Adam
|
| 47 |
Loss Function: Categorical Crossentropy
|
| 48 |
Metrics: Accuracy
|
|
|
|
| 49 |
Training
|
|
|
|
| 50 |
Epochs: 10
|
|
|
|
| 51 |
Steps per Epoch: Calculated based on the training dataset size and batch size.
|
| 52 |
Validation Steps: Calculated based on the validation dataset size and batch size.
|
| 53 |
Model Save
|
|
|
|
| 10 |
- ML
|
| 11 |
- Ai
|
| 12 |
---
|
| 13 |
+
Facial Recognition Model
|
| 14 |
+
|
| 15 |
Model Information
|
| 16 |
+
|
| 17 |
Name: Facial Means
|
| 18 |
+
|
| 19 |
Type: Convolutional Neural Network (CNN)
|
| 20 |
+
|
| 21 |
Framework: TensorFlow
|
| 22 |
+
|
| 23 |
Dataset: Celebrity Faces Dataset
|
| 24 |
+
|
| 25 |
Dataset Path: /content/drive/MyDrive/beard_dataset/celb_dataset/
|
| 26 |
+
|
| 27 |
Model Save Path: /content/drive/MyDrive/beard_dataset/celebrity_model.h5
|
| 28 |
+
|
| 29 |
Image Dimensions: 224 x 224 pixels
|
| 30 |
+
|
| 31 |
Batch Size: 32
|
| 32 |
Data Augmentation
|
| 33 |
Rescale: 1./255
|
|
|
|
| 36 |
Horizontal Flip: True
|
| 37 |
Validation Split: 20%
|
| 38 |
Model Architecture
|
| 39 |
+
Layer (type)
|
| 40 |
+
Output Shape Param #
|
| 41 |
===============================================================
|
| 42 |
conv2d (Conv2D) (None, 222, 222, 32) 896
|
| 43 |
max_pooling2d (MaxPooling2D) (None, 111, 111, 32) 0
|
|
|
|
| 49 |
dense (Dense) (None, 128) 11075712
|
| 50 |
dense_1 (Dense) (None, 6) 774
|
| 51 |
===============================================================
|
| 52 |
+
|
| 53 |
Total params: 11,170,734
|
| 54 |
Trainable params: 11,170,734
|
| 55 |
Non-trainable params: 0
|
|
|
|
| 57 |
Optimizer: Adam
|
| 58 |
Loss Function: Categorical Crossentropy
|
| 59 |
Metrics: Accuracy
|
| 60 |
+
|
| 61 |
Training
|
| 62 |
+
|
| 63 |
Epochs: 10
|
| 64 |
+
|
| 65 |
Steps per Epoch: Calculated based on the training dataset size and batch size.
|
| 66 |
Validation Steps: Calculated based on the validation dataset size and batch size.
|
| 67 |
Model Save
|